1kVA UPS Block Diagram

[fethiyeli]

1kVA Online UPS Block Diagram

View attachment 1kVA UPS Block Diagram.jpg

Hi, i would like to implement a new UPS that includes selected topologies, converters as shown in figure. I will use this block diagram if it is worth to implement. I would like to know whether it is a good design or not. Ofc there are advantages and disadvantages of this design. If this design has any disadvantages, pls share it with me. I really need your advices.

 

[goldsmith]

Hi, i would like to implement a new UPS that includes selected topologies, converters as shown in figure. I will use this block diagram if it is worth to implement. I would like to know whether it is a good design or not. Ofc there are advantages and disadvantages of this design. If this design has any disadvantages, pls share it with me. I really need your advices.

Hi fethiyeli
It seems fine and you can go though it however i won't suggest the buck converter which has been used before the latest stage .
By the way have you ever design any kind of switch mode converter ? or perhaps it is your first design ?
Best Wishes
Goldsmith

 

[fethiyeli]

Hi fethiyeli
It seems fine and you can go though it however i won't suggest the buck converter which has been used before the latest stage .
By the way have you ever design any kind of switch mode converter ? or perhaps it is your first design ?
Best Wishes
Goldsmith

Why dont u suggest the buck converter ? Because of non isolation ? I have to get 311VDC for h-bridge of inverter. I have implemented SMPS before. This is my second design. What topology can u suggest instead of buck converter ? By the way, 30VDC is minimum voltage of three batteries. I have also low voltage cut-off for batteries between flyback converter and full bridge converter.

Thnx.

 

[FvM]

Why dont u suggest the buck converter ? Because of non isolation ? I have to get 311VDC for h-bridge of inverter.

Not necessarily. You can operate the H-bridge with 350 to 390 V bus supply directly and adjust the modulation for constant output voltage.

 

[fethiyeli]

Not necessarily. You can operate the H-bridge with 350 to 390 V bus supply directly and adjust the modulation for constant output voltage.

Yes you are right but i dont have any experience with adjusting width of gate pulses of TL494 via feedback. In fact, ı have used TL494 before for inverter. Which IC can u suggest for inverter ? Any disadvantage or unnecessary converter in this design ?

Thnx.

 

[FvM]

You didn't mention a modulation type of the output inverter. 310V suggests that it is either "modified sine" (= square wave with reduced duty cycle) or SPWM. Changing the modulation index has different effects in both cases.

 

[fethiyeli]

You didn't mention a modulation type of the output inverter. 310V suggests that it is either "modified sine" (= square wave with reduced duty cycle) or SPWM. Changing the modulation index has different effects in both cases.

Yeah but it was mentioned on block diagram =) So dont u suggest changing the duty cycle of PWM ? I guess it can cause distortion of output AC signal. Any other recommend for this block diagram ?

 

[FvM]

Yeah but it was mentioned on block diagram.

So what is it? "220V 50 Hz" doesn't specify a modulation type.

 

[fethiyeli]

So what is it? "220V 50 Hz" doesn't specify a modulation type.

Why dont u just look at the left of the 220V 50Hz ? Just look at the block diagram. It says as PWM inverter. So what do u recommend for this design instead of buck converter ?

 

[FvM]

Why dont u just look at the left of the 220V 50Hz ? Just look at the block diagram. It says as PWM inverter.

And "PWM inverter" means exactly what?

Literally, the term PWM states that there is a modulation involved, so it's not unmodulated square wave or "modified sine". Usually we would expect sine modulation or SPWM (sine PWM). But you also mentioned "adjusting TL494 gate pulses", which doesn't well fit a sine PWM design.

In a sine PWM inverter, the output pulse width is continuously varied between 0 and 100 % duty cyle, so it's easy to superimpose (by multiplication) an additional output voltage adjustment. That's what I suggested in post #4.

 

[fethiyeli]

And "PWM inverter" means exactly what?

Literally, the term PWM states that there is a modulation involved, so it's not unmodulated square wave or "modified sine". Usually we would expect sine modulation or SPWM (sine PWM). But you also mentioned "adjusting TL494 gate pulses", which doesn't well fit a sine PWM design.

In a sine PWM inverter, the output pulse width is continuously varied between 0 and 100 % duty cyle, so it's easy to superimpose (by multiplication) an additional output voltage adjustment. That's what I suggested in post #4.

Yes this inverter is SPWM inverter that uses TL494. I have used half wave inverted AC reference signal in order to get SPWM signals with varying duty cycle. I am not good at electronics as much as you. So, should i use buck converter or just inverter with varying duty cycle respect to output ? TL494 allows 50% duty cycle as maximum.

 

[FvM]

Thanks for clarifying about SPWM. TL494 is designed for push-pull operation with symmetrical duty cycle. Maximum duty cycle is limited by internal dead time to 45..49% for each output, but it can be doubled by paralleling both outputs. Nevertheless the chip isn't well suited for sine PWM, I think.

Adjustment of output level can be done by scaling the sine refernce.

 

[picgak]

Instead of using a buck converter a chopper with control is enough to get 311V after rectification

 

[fethiyeli]

Instead of using a buck converter a chopper with control is enough to get 311V after rectification

How can it be done with DC chopper with control ? I have just heard about DC choppers but not enought to implement in this design.

 

[schmitt trigger]

Please remember, any additional power converter stage that you add, will also add:
1) Cost
2) Size and weight
3) Complexity
4) Efficiency penalty
5) Lower reliability

The relationship 311/220 = 1.41 only applies to a sinewave. For a squarewave, tha ratio is 1.00, which means that your output will be 311 V AC rms (a little less actually, due to losses)

Therefore you have to modulate the inverter...and if you do, then you can use the 390 V DC, no need to step it down.

 

[fethiyeli]

Yes i know them. You re right. My inverter works as follows:

**broken link removed**

**broken link removed**

Grectinvsine is inverted AC signal like image above.

 

[schmitt trigger]

fethiyeli;

I tried to open up your attachments, but it returns an "invalid attachment" error message.

 

[fethiyeli]

 

[schmitt trigger]

Ok, so you'll use SPWM...therefore you can use 390VDC directly, no need to step it down.

 

[fethiyeli]

Ok, so you'll use SPWM...therefore you can use 390VDC directly, no need to step it down.

Exactly, i will use SPWM but i dont know how to change the pulses according to output DC voltage. Output voltage will down to 350VDC but pulse width of the TL494 will remain same since it only varies by inverted AC signal. I guess i need varying pulse width according output voltage when 350VDC is used. If i cant change width of the pulses according to DC bus, output AC voltage will be lower i think. Should i use voltage divider at pin 2 of the TL494 from DC bus. I only know to use it for SMPS design not inverter =) I also dont know which IC is best for inverter design. Other issue is that should i reach 100% duty cycle in order to get 220VAC from 311VDC bus ? I have lots of questions to ask =)